SOFTWARE SIMULATOR FOR ROBOTIC EXPLORATION OF THE LUNAR SURFACE
- Paper number
IAC-09.A3.I.11
- Author
Mr. Riccardo Lombardi, Italy
- Coauthor
Dr. Michèle Lavagna, Politecnico di Milano, Italy
- Coauthor
Mr. Pietro Francesconi, Politecnico di Milano, Italy
- Year
2009
- Abstract
Validation and verification are key points within a space project to check for reliability of the designed product. Of course, those phases are fundamental within the development of projects focused on planetary exploration too, such as those involving mobile elements on the surface; on board software, operations, and operational modes must be checked in a specific simulation environment, possibly including some software representation of the on board hardware too. A space mission simulator allows checking for the design robustness too, as different and complex faulty scenarios may be analysed either to revise the hardware solutions or to correctly define the reconfiguration procedures. Although spacecraft simulators are routine at control centers, rover virtual simulator is a young area for Europe as no space rover has been deployed yet. While the ExoMars program offers a great field for Martian applications, the Lunar Google XPrize challenge represents a perfect lunar scenario to start working on that topic: according to the challenge statement, a rover must be sent on the Moon within 2012, built exploiting only private sponsorships. Italy is running the race with the so called Team Italia: a mainly academic consortium supported by three important space involved national companies. To support the project, the Aerospace Engineering Department of Politecnico di Milano tailored an under development virtual simulator on the lunar mission scenario: at this early phase of the project the robustness of the sensors\actuators architecture can be checked; to visualize the rover behaviour on the surface greatly helps the designers revising and refining the preliminary choices. Moreover, activity planning, operation sequences and operational modes can be easier defined and cross checked. The effects of a lot of faulty conditions may be analyzed to better understand how to cope with either through the design or during operational phases. The simulator here presented supplies: the representation of our satellite environment; orbital mechanics features to let the user identify the visibility windows between the rover antenna and a selected ground station on Earth; representation of the rover – user defined - according to the current design; a navigation module, to define the safe path according to the current map of the environment surrounding the rover; the multi-body model of the rover integrated in the simulation environment, including the traction control of the wheels. The drivers for designing the simulator architecture and selecting the implementation environment for the tool were the portability and the flexibility. The simulator, in fact, must provide a software framework for testing different levels of autonomy - plug-in of external modules and tools must be easy - simulated hardware must be simply redefineable to support the system design process; validation of different granularity of the mission plan shall be feasible The proposed tool merges a physic and a graphic (Irrlicht) engine. Irrlicht is a cross-platform high performance real-time 3D engine; it is a high level API for creating complete 3D and 2D applications like games or scientific visualizations. It comes with a wide documentation and integrates all the state-of-the-art features for visual representation like dynamic shadows, particle systems, character animation, indoor and outdoor technology. The engine is open source. The virtual environment contains all physical properties of the planetary surface, rock distribution and terrain properties, exploited to evaluate the interaction with the rover. The map of the soil properties is an output of the tool, the user can select the soil type, introducing the soil parameters or a predefined soil class; currently the data of Lunar and Martian surface are available, but the tool is highly flexible and configurable. The tool exploits those inputs to build the soil map. The navigation module solves the path-planning problem by applying a potential field method, the target position attracts the rover while obstacles repulse it; therefore, the global effects on the rover, evaluated at each step, determines the future travelling direction. The rover simulator is going to be exploited by the Team Italia to check for rover operations feasibility and robustness both in nominal and possibly faulty scenarios and to validate the design before starting integrating the whole system. The paper will deeply describe the features of the proposed tool and will discuss open points and future refinements.
- Abstract document
- Manuscript document
IAC-09.A3.I.11.pdf (🔒 authorized access only).
To get the manuscript, please contact IAF Secretariat.